A DAG-based sparse Cholesky solver for multicore architectures

ePubs

The open archive for STFC research publications

DOI	10.5286/raltr.2009004
Persistent URL	http://purl.org/net/epubs/work/50205
Record Status	Checked
Record Id	50205
Title	A DAG-based sparse Cholesky solver for multicore architectures
Contributors	JD Hogg (STFC Rutherford Appleton Lab.), JK Reid (STFC Rutherford Appleton Lab.), JA Scott (STFC Rutherford Appleton Lab.)
Abstract	In this paper, we describe the design and development of a new code for the solution of sparse symmetric positive-definite linear systems aimed primarily at multicore architectures. Our new Fortran 95/OpenMP code, HSL_MA67, is available as part of the sofware library HSL and extends to the sparse case the task DAG-based approach that is becoming popular in the design and development of dense linear algebra kernels. Comparisons are made with existing parallel solvers, using problems arising from a range of practical applications. We demonstrate thta HSL_MA87 obtains good serial and parallel times on an 8-core machine.
Organisation	CSE , CSE-NAG , STFC
Keywords	Fortran 95 , parallel , Cholesky factorization , sparse symmetric linear systems , DAG-based , multicore , OpenMP
Funding Information
Related Research Object(s):	60482 , 60479
Licence Information:
Language	English (EN)

Type	Details	URI(s)	Local file(s)	Year
Report	RAL Technical Reports RAL-TR-2009-004. STFC, 2009.		hrsRALTR2009004.pdf	2009